The accumulation of amyloid [unreadable] peptide (A[unreadable]) within senile plaques and the walls of cerebral blood vessels, as well as diffuse A[unreadable] deposits is a hallmark of Alzheimer's disease (AD) pathology. Most AD-linked mutations precipitate A[unreadable] aggregation leading to its increasing toxicity and dementia. The main goal of this study is to elucidate the relationship between cognitive deficits induced by toxic species of A[unreadable]1-42 in mice. Though animal modelling studies demonstrate that A[unreadable] aggregates likely mediate, at least to some extent, cognitive dysfunction in transgenic amyloid precursor protein (APP) mice, the relationship of this dysfunction to dementia in AD patients remains unclear. We hypothesize that if the mnemonic dysfunction in mice reflects underlying processes in AD dementia, then CNS expression of internal A[unreadable] mutations that cause familial AD (FAD) should also produce a greater degree of cognitive dysfunction in mouse models then expression of wt A[unreadable]. There is a strong scientific rationale for studying "internal" FAD-linked mutant A[unreadable]s in vivo. We have chosen one A[unreadable] mutant that has already been studied in the context of APP Tg mice, E22G ("Arctic"). Including this mutant will enable us to benchmark our findings using a novel method of A[unreadable] expression against existing TgAPP E22G mice. We chose a second recently reported mutant (?E22), based on its novelty and the fact that it was claimed to cause AD without amyloid. We have developed unique technologies that enable us to express individual A[unreadable] peptides in the secretory extracellular pathways or in the intracellular pathways in the brains of mice. It has been proposed that intracellular A[unreadable] causes cognitive deficits in APP transgenic mice. Because A[unreadable] over-expression driven by APP produces both intracellular and extracellular A[unreadable] it is very challenging to prove which form drives the cognitive deficits. By targeting A[unreadable] preferentially into intracellular pathways we should be able to determine if intracellular A[unreadable] results in cognitive impairment. Thus comparing effects of individual A[unreadable] mutants should help to establish the relationship between intrinsic mutations in the A[unreadable] sequence that cause AD and the cognitive deficits in mice. We predict that, if the cognitive deficit is a reliable surrogate for pathologic phenomena associated with AD dementia, a mutant A[unreadable] that causes AD should result in stronger cognitive impairment then the wild-type A[unreadable]. Such targeting studies should provide critical insight into the cascade of events where A[unreadable] exerts its behavioural effects. Our central hypothesis is that if the behavioural deficits studied to date in TgAPP mice are relevant to the study of AD and are good correlates of A[unreadable] pathology, then understanding the mechanisms by which A[unreadable] induces cognitive deficits in mice, may lead to novel therapies aimed at improving cognition in humans. PUBLIC HEALTH RELEVANCE: Highly fibrillogenic and more toxic sub-form of amyloid 2 peptide (A[unreadable]42) is implicated in neuronal death and dementia in Alzheimer's disease. We will use novel technology of expressing various mutations of A[unreadable]42 in mouse brain and we will compare the effects of these individual A[unreadable] mutants on cognition in mice. Our study will potentially provide a novel paradigm for performing model genetic manipulations within the brain that can be analyzed using behavioral assays. The understanding of the mechanisms by which A[unreadable] induces dementia in mice, may lead to development of novel therapies aimed at improving cognition in humans.